In [1]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import csv
import scanpy.api as sc
from igraph import *
from MulticoreTSNE import MulticoreTSNE as TSNE #faster TSNE alternative
from anndata import read_h5ad
import dill # for saving everything

/home/ubuntu/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters

Load data

18, 21 and 24 months

In [2]:
path = '/data/maca/data/'
maca10x182124metadata = pd.read_csv('/data/maca/data/MACA_10x_18-21-24_qc.csv',usecols = ['channel','tissue','tissue.notes','mouse.age','mouse.sex'])

maca10x182124metadata = maca10x182124metadata.rename(columns = {'mouse.age':'age','mouse.sex':'sex','tissue.notes':'subtissue'})
maca10x182124metadata.index = range(len(maca10x182124metadata))
In [3]:
maca10x182124metadata.head()
#len(maca10x182124metadata)
Out[3]:
channel age sex tissue subtissue
0 MACA_18m_F_AORTA_50 18 F AORTA NaN
1 MACA_18m_F_BM_50 18 F MARROW NaN
2 MACA_18m_F_BM_51 18 F MARROW NaN
3 MACA_18m_F_BREAST_50 18 F MAMMARY NaN
4 MACA_18m_F_BREAST_51 18 F MAMMARY NaN
In [4]:
adata182124aux = []
for i in range(0,len(maca10x182124metadata)):
    path = '/data/maca/data/' + maca10x182124metadata.channel[i] + '/'
    foo = sc.read(path + 'matrix.mtx', cache=True).transpose()
    foo.var_names = np.genfromtxt(path + 'genes.tsv', dtype=str)[:, 1]
    foo.obs_names = np.genfromtxt(path + 'barcodes.tsv', dtype=str)
    foo.obs['Tissue'] = str(maca10x182124metadata.tissue[i]).title()
    #foo.obs['subTissue'] = str(maca10x182124metadata.subtissue[i]).title()
    
    
    if pd.isnull(maca10x182124metadata.subtissue[i]):
        foo.obs['subTissue'] = maca10x182124metadata.subtissue[i]
    else:        
        foo.obs['subTissue'] = str(maca10x182124metadata.subtissue[i]).title()
        
    
    foo.obs['Age'] = str(maca10x182124metadata.age[i]) + 'm'
    
    if maca10x182124metadata.sex[i] == 'M':
        foo.obs['Sex'] = 'male'
    else:
        foo.obs['Sex'] = 'female'
    sc.pp.filter_cells(foo, min_genes=250)
    
    foo.obs['Channel'] = maca10x182124metadata.channel[i]# + foo.obs_names.split("-")[0]
    
    foo.obs['Cell_Ontology_Type'] = 'NA' #maca10x3metadata.cell_ontology_class[i]
    #foo.obs['CellTypeID'] = maca10x3metadata.cell_ontology_id[i]
    #foo.obs['FreeAnn'] = maca10x3metadata.free_annotation[i]
    
    adata182124aux.append(foo)
    #print(adata3.X.size*8/1000000) # convert to Mb


first = adata182124aux.pop()
adata182124 = first.concatenate(adata182124aux)
adata182124.obs['Cell'] = adata182124.obs['Channel'] + '_' + adata182124.obs_names
adata182124.obs['Cell'] = adata182124.obs['Cell'].apply(lambda x: pd.Series(x.split('-')))[0]

adata182124.obs.head()
Out[4]:
Tissue subTissue Age Sex n_genes Channel Cell_Ontology_Type batch Cell
AAACCTGAGGCACATG-1-0 Tongue NaN 24m male 510 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGAGGCACATG
AAACCTGCAGGGTACA-1-0 Tongue NaN 24m male 2107 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGCAGGGTACA
AAACCTGCAGTAAGCG-1-0 Tongue NaN 24m male 3481 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGCAGTAAGCG
AAACCTGTCATTATCC-1-0 Tongue NaN 24m male 2599 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGTCATTATCC
AAACGGGAGATGAGAG-1-0 Tongue NaN 24m male 410 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGAGATGAGAG
In [5]:
adata182124.obs
Out[5]:
Tissue subTissue Age Sex n_genes Channel Cell_Ontology_Type batch Cell
AAACCTGAGGCACATG-1-0 Tongue NaN 24m male 510 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGAGGCACATG
AAACCTGCAGGGTACA-1-0 Tongue NaN 24m male 2107 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGCAGGGTACA
AAACCTGCAGTAAGCG-1-0 Tongue NaN 24m male 3481 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGCAGTAAGCG
AAACCTGTCATTATCC-1-0 Tongue NaN 24m male 2599 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGTCATTATCC
AAACGGGAGATGAGAG-1-0 Tongue NaN 24m male 410 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGAGATGAGAG
AAACGGGGTACAGTGG-1-0 Tongue NaN 24m male 3468 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGGTACAGTGG
AAACGGGGTCAATACC-1-0 Tongue NaN 24m male 537 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGGTCAATACC
AAACGGGGTCTTCTCG-1-0 Tongue NaN 24m male 3189 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGGTCTTCTCG
AAACGGGTCATGTCCC-1-0 Tongue NaN 24m male 504 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGTCATGTCCC
AAAGATGAGCTATGCT-1-0 Tongue NaN 24m male 3419 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGATGAGCTATGCT
AAAGATGAGGTCGGAT-1-0 Tongue NaN 24m male 3764 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGATGAGGTCGGAT
AAAGATGGTCAGAAGC-1-0 Tongue NaN 24m male 276 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGATGGTCAGAAGC
AAAGATGGTGCAGGTA-1-0 Tongue NaN 24m male 3879 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGATGGTGCAGGTA
AAAGCAAAGTGAATTG-1-0 Tongue NaN 24m male 567 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGCAAAGTGAATTG
AAAGCAAGTAGATTAG-1-0 Tongue NaN 24m male 3496 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGCAAGTAGATTAG
AAAGCAAGTATTCTCT-1-0 Tongue NaN 24m male 5439 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGCAAGTATTCTCT
AAAGCAAGTTAAGATG-1-0 Tongue NaN 24m male 1251 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGCAAGTTAAGATG
AAAGTAGAGCGAAGGG-1-0 Tongue NaN 24m male 405 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGAGCGAAGGG
AAAGTAGCAAAGTCAA-1-0 Tongue NaN 24m male 2397 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCAAAGTCAA
AAAGTAGCAAGCCCAC-1-0 Tongue NaN 24m male 327 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCAAGCCCAC
AAAGTAGCACATCTTT-1-0 Tongue NaN 24m male 3970 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCACATCTTT
AAAGTAGCACCTTGTC-1-0 Tongue NaN 24m male 2954 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCACCTTGTC
AAAGTAGCAGGCTCAC-1-0 Tongue NaN 24m male 4764 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCAGGCTCAC
AAAGTAGCAGGTGGAT-1-0 Tongue NaN 24m male 3036 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCAGGTGGAT
AAAGTAGCATCGATTG-1-0 Tongue NaN 24m male 2696 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGCATCGATTG
AAAGTAGGTAGCGCTC-1-0 Tongue NaN 24m male 266 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGGTAGCGCTC
AAAGTAGGTATTCTCT-1-0 Tongue NaN 24m male 2792 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGGTATTCTCT
AAAGTAGGTTGCCTCT-1-0 Tongue NaN 24m male 5198 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGGTTGCCTCT
AAAGTAGTCACCGGGT-1-0 Tongue NaN 24m male 630 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAAGTAGTCACCGGGT
AAATGCCAGCACAGGT-1-0 Tongue NaN 24m male 3693 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAATGCCAGCACAGGT
... ... ... ... ... ... ... ... ... ...
TTTACTGCACGGTGTC-1-89 Tongue NaN 24m male 3274 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTACTGCACGGTGTC
TTTACTGGTCGAAAGC-1-89 Tongue NaN 24m male 3965 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTACTGGTCGAAAGC
TTTACTGTCATCATTC-1-89 Tongue NaN 24m male 1076 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTACTGTCATCATTC
TTTACTGTCGTCTGCT-1-89 Tongue NaN 24m male 2752 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTACTGTCGTCTGCT
TTTATGCAGAGACGAA-1-89 Tongue NaN 24m male 3693 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTATGCAGAGACGAA
TTTATGCAGGCTCATT-1-89 Tongue NaN 24m male 2007 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTATGCAGGCTCATT
TTTATGCCAACTTGAC-1-89 Tongue NaN 24m male 1050 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTATGCCAACTTGAC
TTTATGCCACAAGTAA-1-89 Tongue NaN 24m male 1559 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTATGCCACAAGTAA
TTTATGCCACAGGCCT-1-89 Tongue NaN 24m male 2582 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTATGCCACAGGCCT
TTTATGCGTCGCGAAA-1-89 Tongue NaN 24m male 382 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTATGCGTCGCGAAA
TTTCCTCAGCAGCGTA-1-89 Tongue NaN 24m male 4550 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTCCTCAGCAGCGTA
TTTCCTCAGTACATGA-1-89 Tongue NaN 24m male 966 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTCCTCAGTACATGA
TTTCCTCCAGTTTACG-1-89 Tongue NaN 24m male 3629 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTCCTCCAGTTTACG
TTTCCTCGTCACCCAG-1-89 Tongue NaN 24m male 447 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTCCTCGTCACCCAG
TTTCCTCGTGTAACGG-1-89 Tongue NaN 24m male 2732 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTCCTCGTGTAACGG
TTTCCTCTCACTTCAT-1-89 Tongue NaN 24m male 3877 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTCCTCTCACTTCAT
TTTGCGCAGGAATTAC-1-89 Tongue NaN 24m male 308 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGCGCAGGAATTAC
TTTGCGCCAATGAAAC-1-89 Tongue NaN 24m male 825 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGCGCCAATGAAAC
TTTGCGCGTTGGGACA-1-89 Tongue NaN 24m male 483 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGCGCGTTGGGACA
TTTGCGCTCATTCACT-1-89 Tongue NaN 24m male 271 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGCGCTCATTCACT
TTTGCGCTCCTGCAGG-1-89 Tongue NaN 24m male 1813 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGCGCTCCTGCAGG
TTTGGTTTCAACGGCC-1-89 Tongue NaN 24m male 2612 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGGTTTCAACGGCC
TTTGGTTTCACTTCAT-1-89 Tongue NaN 24m male 296 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGGTTTCACTTCAT
TTTGGTTTCGTTGCCT-1-89 Tongue NaN 24m male 2465 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGGTTTCGTTGCCT
TTTGGTTTCTAAGCCA-1-89 Tongue NaN 24m male 2998 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGGTTTCTAAGCCA
TTTGTCAAGCGGATCA-1-89 Tongue NaN 24m male 4521 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGTCAAGCGGATCA
TTTGTCACAATGTTGC-1-89 Tongue NaN 24m male 1647 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGTCACAATGTTGC
TTTGTCAGTCATCCCT-1-89 Tongue NaN 24m male 4784 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGTCAGTCATCCCT
TTTGTCAGTCTACCTC-1-89 Tongue NaN 24m male 3280 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGTCAGTCTACCTC
TTTGTCATCTGCGACG-1-89 Tongue NaN 24m male 3092 MACA_24m_M_TONGUE_59 NA 89 MACA_24m_M_TONGUE_59_TTTGTCATCTGCGACG

220647 rows × 9 columns

3months

In [46]:
path = '/data/maca/data/10x/'
maca10x3metadata = pd.read_csv('/data/maca/data/10x/MACA_10x_3_qc.csv',usecols = ['channel','tissue','tissue.notes','mouse.age','mouse.sex','pool'])
maca10x3metadata2 = pd.read_csv('/data/maca/data/10x/TM_droplet_metadata.csv', low_memory=False)

maca10x3metadata = maca10x3metadata.rename(columns = {'mouse.age':'age','mouse.sex':'sex','tissue.notes':'subtissue'})
#maca10x3metadata = maca10x3metadata[maca10x3metadata.age == 3]
maca10x3metadata.index = range(len(maca10x3metadata))

maca10x3metadata.head()
Out[46]:
channel tissue subtissue age sex pool
0 10X_P4_0 Tongue TONGUE 3 M 4
1 10X_P4_1 Tongue TONGUE 3 M 4
2 10X_P4_2 Liver HEPATOCYTES 3 M 4
3 10X_P4_3 Bladder BLADDER 3 M 4
4 10X_P4_4 Bladder BLADDER 3 M 4
In [47]:
adata3aux = []
for i in range(0,len(maca10x3metadata)):
    #print(i)
    path = '/data/maca/data/10x/' + maca10x3metadata.channel[i] + '/'
    foo = sc.read(path + 'matrix.mtx', cache=True).transpose()
    foo.var_names = np.genfromtxt(path + 'genes.tsv', dtype=str)[:, 1]
    foo.obs_names = np.genfromtxt(path + 'barcodes.tsv', dtype=str)
    foo.obs['Tissue'] = maca10x3metadata.tissue[i]
    foo.obs['subTissue'] = maca10x3metadata.subtissue[i]
    foo.obs['Age'] = str(maca10x3metadata.age[i]) + 'm'
    
    if maca10x3metadata.sex[i] == 'M':
        foo.obs['Sex'] = 'male'
    else:
        foo.obs['Sex'] = 'female'
    sc.pp.filter_cells(foo, min_genes=250)
    
    foo.obs['Channel'] = maca10x3metadata.channel[i]# + foo.obs_names.split("-")[0]
    
    foo.obs['Cell_Ontology_Type'] = 'NA' #maca10x3metadata.cell_ontology_class[i]
    #foo.obs['CellTypeID'] = maca10x3metadata.cell_ontology_id[i]
    #foo.obs['FreeAnn'] = maca10x3metadata.free_annotation[i]
    
    adata3aux.append(foo)
    #print(adata3.X.size*8/1000000) # convert to Mb

first = adata3aux.pop()
adata3 = first.concatenate(adata3aux)
adata3.obs['Cell'] = adata3.obs['Channel'] + '_' + adata3.obs_names
adata3.obs['Cell'] = adata3.obs['Cell'].apply(lambda x: pd.Series(x.split('-')))[0]

display(adata3.obs.head())
adata3
Tissue subTissue Age Sex n_genes Channel Cell_Ontology_Type batch Cell
AAACCTGAGAGCCCAA-1-0 Trachea NaN 3m male 669 10X_P8_15 NA 0 10X_P8_15_AAACCTGAGAGCCCAA
AAACCTGAGCTAGTCT-1-0 Trachea NaN 3m male 1071 10X_P8_15 NA 0 10X_P8_15_AAACCTGAGCTAGTCT
AAACCTGAGGATGTAT-1-0 Trachea NaN 3m male 1311 10X_P8_15 NA 0 10X_P8_15_AAACCTGAGGATGTAT
AAACCTGAGGGATACC-1-0 Trachea NaN 3m male 1804 10X_P8_15 NA 0 10X_P8_15_AAACCTGAGGGATACC
AAACCTGAGTTTAGGA-1-0 Trachea NaN 3m male 1085 10X_P8_15 NA 0 10X_P8_15_AAACCTGAGTTTAGGA
Out[47]:
AnnData object with n_obs × n_vars = 58510 × 23433 
    obs: 'Tissue', 'subTissue', 'Age', 'Sex', 'n_genes', 'Channel', 'Cell_Ontology_Type', 'batch', 'Cell'

concatenate all time points

In [48]:
adata = adata182124.concatenate([adata3])
adata.write('./write/maca.h5ad')
In [ ]:
import gc
gc.collect()

Import all data

In [ ]:
adata = read_h5ad('./write/maca.h5ad')
type(adata.X)
In [49]:
adata.shape
Out[49]:
(279157, 23433)
In [50]:
adata
Out[50]:
AnnData object with n_obs × n_vars = 279157 × 23433 
    obs: 'Tissue', 'subTissue', 'Age', 'Sex', 'n_genes', 'Channel', 'Cell_Ontology_Type', 'batch', 'Cell'
In [51]:
adata.obs.head()
Out[51]:
Tissue subTissue Age Sex n_genes Channel Cell_Ontology_Type batch Cell
AAACCTGAGGCACATG-1-0-0 Tongue NaN 24m male 510 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGAGGCACATG
AAACCTGCAGGGTACA-1-0-0 Tongue NaN 24m male 2107 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGCAGGGTACA
AAACCTGCAGTAAGCG-1-0-0 Tongue NaN 24m male 3481 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGCAGTAAGCG
AAACCTGTCATTATCC-1-0-0 Tongue NaN 24m male 2599 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACCTGTCATTATCC
AAACGGGAGATGAGAG-1-0-0 Tongue NaN 24m male 410 MACA_24m_M_TONGUE_60 NA 0 MACA_24m_M_TONGUE_60_AAACGGGAGATGAGAG

Preprocessing

In [52]:
sc.pp.filter_genes(adata, min_cells=5)
sc.pp.filter_cells(adata, min_genes=250)
adata
Out[52]:
AnnData object with n_obs × n_vars = 279156 × 19614 
    obs: 'Tissue', 'subTissue', 'Age', 'Sex', 'n_genes', 'Channel', 'Cell_Ontology_Type', 'batch', 'Cell'
    var: 'n_cells'
In [53]:
# add the total counts per cell as observations-annotation to adata
adata.obs['n_counts'] = np.sum(adata.X, axis=1).A1
adata
Out[53]:
AnnData object with n_obs × n_vars = 279156 × 19614 
    obs: 'Tissue', 'subTissue', 'Age', 'Sex', 'n_genes', 'Channel', 'Cell_Ontology_Type', 'batch', 'Cell', 'n_counts'
    var: 'n_cells'
In [54]:
axs = sc.pl.violin(adata, ['n_genes', 'n_counts'],
                   jitter=0.4, multi_panel=True)

... storing 'Tissue' as categorical
... storing 'subTissue' as categorical
... storing 'Age' as categorical
... storing 'Sex' as categorical
... storing 'Channel' as categorical
... storing 'Cell_Ontology_Type' as categorical
In [55]:
ax = sc.pl.scatter(adata, x='n_counts', y='n_genes')
In [56]:
sc.pp.normalize_per_cell(adata, counts_per_cell_after=1e4) #simple lib size normalization?
adata.raw = adata
In [57]:
filter_result = sc.pp.filter_genes_dispersion(
    adata.X, min_mean=0.0125, max_mean=10, min_disp=0.5)
sc.pl.filter_genes_dispersion(filter_result)
In [58]:
adata = adata[:, filter_result.gene_subset]
adata.shape
Out[58]:
(279156, 2202)
In [59]:
sc.pp.log1p(adata)
In [60]:
sc.pp.scale(adata, max_value=10, zero_center=False)

Exploration

PCA

In [61]:
sc.tl.pca(adata)
In [62]:
ax = sc.pl.pca_scatter(adata, color=['Tissue'], right_margin=0.5)
In [63]:
ax = sc.pl.pca_scatter(adata, color=['Age'], right_margin=0.5)
In [64]:
ax = sc.pl.pca_scatter(adata, color=['Sex'], right_margin=0.5)
In [65]:
ax = sc.pl.pca_scatter(adata, color='n_counts', right_margin=0.5)
In [66]:
sc.pl.pca_variance_ratio(adata, log=True)

Louvain clustering

In [67]:
sc.pp.neighbors(adata, n_neighbors=15)#, method='gauss')
sc.tl.louvain(adata, resolution = 0.3)
In [68]:
adata
Out[68]:
AnnData object with n_obs × n_vars = 279156 × 2202 
    obs: 'Tissue', 'subTissue', 'Age', 'Sex', 'n_genes', 'Channel', 'Cell_Ontology_Type', 'batch', 'Cell', 'n_counts', 'louvain'
    var: 'n_cells'
    uns: 'pca', 'Tissue_colors', 'Age_colors', 'Sex_colors', 'neighbors', 'louvain'
    obsm: 'X_pca'
    varm: 'PCs'

UMAP

In [69]:
sc.tl.umap(adata)
In [80]:
sc.settings.set_figure_params(dpi=200)
sc.pl.umap(adata, color=['Tissue'], right_margin=0.5, save='_all_ages_tissue.pdf')

saving figure to file ./figures/umap_all_ages_tissue.pdf
In [81]:
sc.settings.set_figure_params(dpi=200)
sc.pl.umap(adata, color=['Age'], right_margin=0.5, save='_all_ages_age.pdf')

saving figure to file ./figures/umap_all_ages_age.pdf
In [87]:
sc.settings.set_figure_params(dpi=200)
sc.pl.umap(adata, color=['Sex'], right_margin=0.5, save='_all_ages_sex.pdf')

saving figure to file ./figures/umap_all_ages_sex.pdf
In [82]:
sc.settings.set_figure_params(dpi=200)
sc.pl.umap(adata, color=['louvain'], right_margin=0.5, save='_all_ages_louvain.pdf')

saving figure to file ./figures/umap_all_ages_louvain.pdf

tSNE

In [73]:
sc.tl.tsne(adata, perplexity=50)
In [83]:
sc.settings.set_figure_params(dpi=200)
sc.pl.tsne(adata, color=['Tissue'], right_margin=0.5, save='_all_ages_tissue.pdf')

saving figure to file ./figures/tsne_all_ages_tissue.pdf
In [86]:
sc.pl.tsne(adata, color=['Age'], right_margin=0.5, save='_all_ages_age.pdf')

saving figure to file ./figures/tsne_all_ages_age.pdf
In [84]:
sc.pl.tsne(adata, color=['Sex'], right_margin=0.5, save='_all_ages_sex.pdf')

saving figure to file ./figures/tsne_all_ages_sex.pdf
In [85]:
sc.pl.tsne(adata, color=['louvain'], right_margin=0.5, save='_all_ages_louvain.pdf')

saving figure to file ./figures/tsne_all_ages_louvain.pdf

Save processed data

In [88]:
adata.write('./write/maca.processed.h5ad')